Mistral: Mistral Nemo

Text input Text output Free Option
Author's Description

A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, and Hindi. It supports function calling and is released under the Apache 2.0 license.

Key Specifications
Cost
$
Context
131K
Parameters
12B (Rumoured)
Released
Jul 18, 2024
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Response Format Stop Seed Min P Top P Max Tokens Frequency Penalty Temperature Presence Penalty
Features

This model supports the following features:

Response Format
Performance Summary

Mistral Nemo, a 12B parameter multilingual model developed by Mistral in collaboration with NVIDIA, demonstrates exceptional speed and competitive pricing. It consistently ranks among the fastest models, achieving an Infinityth percentile across nine benchmarks, and offers highly competitive pricing, ranking in the 97th percentile across eight benchmarks. The model also exhibits strong reliability with a 91% success rate. In terms of benchmark performance, Mistral Nemo excels in Ethics, achieving perfect 100% accuracy and earning accolades for being the most accurate model at its price point, fastest among models of comparable accuracy, and having the best accuracy-to-cost ratio. It also shows a strong performance in Coding (79.0% accuracy) and Email Classification (93.0% accuracy). However, the model struggles significantly with Instruction Following, showing 0.0% accuracy in one test and 37.0% in another, and performs poorly in Mathematics (15.0% accuracy) and Reasoning (28.0% accuracy). Its hallucination rate is moderate at 62.0% accuracy, indicating room for improvement in acknowledging uncertainty. Despite some weaknesses, its multilingual support and function calling capabilities, combined with its speed and cost-effectiveness, make it a compelling option for specific applications.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.02
Completion $0.04

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
DeepInfra
DeepInfra | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.04 / 1M tokens
Kluster
Kluster | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.04 / 1M tokens
Enfer
Enfer | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.04 / 1M tokens
Parasail
Parasail | mistralai/mistral-nemo 131K $0.03 / 1M tokens $0.11 / 1M tokens
NextBit
NextBit | mistralai/mistral-nemo 128K $0.02 / 1M tokens $0.04 / 1M tokens
InferenceNet
InferenceNet | mistralai/mistral-nemo 16K $0.0375 / 1M tokens $0.1 / 1M tokens
Nebius
Nebius | mistralai/mistral-nemo 128K $0.02 / 1M tokens $0.04 / 1M tokens
Novita
Novita | mistralai/mistral-nemo 60K $0.04 / 1M tokens $0.17 / 1M tokens
Atoma
Atoma | mistralai/mistral-nemo 128K $0.02 / 1M tokens $0.04 / 1M tokens
InoCloud
InoCloud | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.04 / 1M tokens
Mistral
Mistral | mistralai/mistral-nemo 131K $0.15 / 1M tokens $0.15 / 1M tokens
Azure
Azure | mistralai/mistral-nemo 128K $0.3 / 1M tokens $0.3 / 1M tokens
Enfer
Enfer | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.04 / 1M tokens
Nineteen
Nineteen | mistralai/mistral-nemo 32K $0.02 / 1M tokens $0.04 / 1M tokens
Chutes
Chutes | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.04 / 1M tokens
Chutes
Chutes | mistralai/mistral-nemo 131K $0.02 / 1M tokens $0.07 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by mistralai